notebooks/05_Embeddings_in_Transformers.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true,
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "85962e48f2d2442cba4151bd6930ae9c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_8dd9f660ef2e44b9bfa7a6261ac08ba7",
              "IPY_MODEL_4794a0a52f564f13ae3882c3b3060e8d",
              "IPY_MODEL_0d90e6d600284ee597f7d6085f5fafb8"
            ],
            "layout": "IPY_MODEL_e453a5aa43f7442f8f36c9609c0c0e8c"
          }
        },
        "8dd9f660ef2e44b9bfa7a6261ac08ba7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2cfeb1677ec24e078a67b88a68a0f417",
            "placeholder": "​",
            "style": "IPY_MODEL_ef0b8c38eda24ef2beb4fafab3803947",
            "value": "Downloading (…)solve/main/vocab.txt: 100%"
          }
        },
        "4794a0a52f564f13ae3882c3b3060e8d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d686d880344a4451bbec22f5d59933c2",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_573e47ae4244410bb7d94e9254765e9b",
            "value": 231508
          }
        },
        "0d90e6d600284ee597f7d6085f5fafb8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c1a998ccdae948158944404c564e4c39",
            "placeholder": "​",
            "style": "IPY_MODEL_43d7325199d2458da0809f72282c5d9e",
            "value": " 232k/232k [00:00&lt;00:00, 3.22MB/s]"
          }
        },
        "e453a5aa43f7442f8f36c9609c0c0e8c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2cfeb1677ec24e078a67b88a68a0f417": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ef0b8c38eda24ef2beb4fafab3803947": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d686d880344a4451bbec22f5d59933c2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "573e47ae4244410bb7d94e9254765e9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "c1a998ccdae948158944404c564e4c39": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "43d7325199d2458da0809f72282c5d9e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "2b94ad1e89e048f0a6121eb352f1db7f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9a2dea225e6845dfafa19e45aaf5dbfe",
              "IPY_MODEL_3862e7c4587f412981e5942a9dc9f6cb",
              "IPY_MODEL_346f11c6031641b082a109f2c45da0d3"
            ],
            "layout": "IPY_MODEL_bec22b9870674282af9797929ea154c3"
          }
        },
        "9a2dea225e6845dfafa19e45aaf5dbfe": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b9d38c87444949c399206f644804f6fd",
            "placeholder": "​",
            "style": "IPY_MODEL_696a97128cc342738076343c298bbee9",
            "value": "Downloading (…)okenizer_config.json: 100%"
          }
        },
        "3862e7c4587f412981e5942a9dc9f6cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_26d9d1914b874de1ba68a3eb3881c4a5",
            "max": 28,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_8adcf7b7db6b4e679505e848ef651a5f",
            "value": 28
          }
        },
        "346f11c6031641b082a109f2c45da0d3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d8f401d1173c43b4bae3c8c1a3214b84",
            "placeholder": "​",
            "style": "IPY_MODEL_34e30849de5948208424f4900a9dfffa",
            "value": " 28.0/28.0 [00:00&lt;00:00, 779B/s]"
          }
        },
        "bec22b9870674282af9797929ea154c3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b9d38c87444949c399206f644804f6fd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "696a97128cc342738076343c298bbee9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "26d9d1914b874de1ba68a3eb3881c4a5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8adcf7b7db6b4e679505e848ef651a5f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "d8f401d1173c43b4bae3c8c1a3214b84": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "34e30849de5948208424f4900a9dfffa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cd08738f01a441fea32b06803a12ee1f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2317a7610b444efc901270a2ae5ff32a",
              "IPY_MODEL_f377afae00fa4ec0b5283ef87cd5da9b",
              "IPY_MODEL_a82ae221d0434d8aad909d7309d432ad"
            ],
            "layout": "IPY_MODEL_169a5aea8df04a66a2202fe329dd8d2a"
          }
        },
        "2317a7610b444efc901270a2ae5ff32a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8995160597e540628dc4fd94420d431b",
            "placeholder": "​",
            "style": "IPY_MODEL_a2450d05a87e4fcea27e5aca1dd7cf3f",
            "value": "Downloading (…)lve/main/config.json: 100%"
          }
        },
        "f377afae00fa4ec0b5283ef87cd5da9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_12f72f729d7a410788e58e0acce5c3ff",
            "max": 570,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_51ddb379365c4aae9414afee809816c5",
            "value": 570
          }
        },
        "a82ae221d0434d8aad909d7309d432ad": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d9ef77507a134c25b7368087c84974ff",
            "placeholder": "​",
            "style": "IPY_MODEL_0570e6d2f3594afa8851e1ff0ff816c2",
            "value": " 570/570 [00:00&lt;00:00, 20.8kB/s]"
          }
        },
        "169a5aea8df04a66a2202fe329dd8d2a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8995160597e540628dc4fd94420d431b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a2450d05a87e4fcea27e5aca1dd7cf3f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "12f72f729d7a410788e58e0acce5c3ff": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "51ddb379365c4aae9414afee809816c5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "d9ef77507a134c25b7368087c84974ff": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0570e6d2f3594afa8851e1ff0ff816c2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a859d97a8bce49d7b32b18f38e58dfe7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_f2937a141b39436dadf80f8a1767e239",
              "IPY_MODEL_cf3b47a516db4c728212891b4bd0a1f3",
              "IPY_MODEL_b9ac95eff849486c805af6d358e019eb"
            ],
            "layout": "IPY_MODEL_a156cd35bfb94ec48b21ddb99341bc77"
          }
        },
        "f2937a141b39436dadf80f8a1767e239": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_17c22bc6de5b486abcb312289830d4b0",
            "placeholder": "​",
            "style": "IPY_MODEL_fc9459527210450c8e00301f37384ee9",
            "value": "Downloading model.safetensors: 100%"
          }
        },
        "cf3b47a516db4c728212891b4bd0a1f3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e176ff18d3764443bf200b72df1a7340",
            "max": 440449768,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_c8dbcd3d362d48f89c97f40b84b86d65",
            "value": 440449768
          }
        },
        "b9ac95eff849486c805af6d358e019eb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0af689241e7848a780158989c68ab7aa",
            "placeholder": "​",
            "style": "IPY_MODEL_1233f7a9d7e642799a0307187ec64b2d",
            "value": " 440M/440M [00:03&lt;00:00, 110MB/s]"
          }
        },
        "a156cd35bfb94ec48b21ddb99341bc77": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "17c22bc6de5b486abcb312289830d4b0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fc9459527210450c8e00301f37384ee9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e176ff18d3764443bf200b72df1a7340": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c8dbcd3d362d48f89c97f40b84b86d65": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "0af689241e7848a780158989c68ab7aa": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1233f7a9d7e642799a0307187ec64b2d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "Transformers have revolutionized the world of Natural Language Processing (NLP) with their immense capacity to understand and generate human-like text. Central to this capability are embeddings, which efficiently convert text data into numerical representations. In this comprehensive guide, we'll dive deep into the world of embeddings in transformers, employing PyTorch and the HuggingFace library.\n",
        "\n",
        "##1. Introduction\n",
        "In NLP, embeddings transition text into a numerical realm, retaining semantics. This numerical representation is essential for feeding text into neural networks. Transformers employ multiple embeddings, including token, positional, and segment embeddings. Their combined might empowers transformers to understand text nuances and order."
      ],
      "metadata": {
        "id": "GDNpVyZLv8p5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##2. Prerequisites: Setting Up the Environment\n",
        "Before diving deep, ensure you've set up your environment:"
      ],
      "metadata": {
        "id": "cFgt_XwFwDXe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -q torch\n",
        "!pip install -q transformers"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EXZuLkjug11m",
        "outputId": "23ae8a29-0dde-4790-aa5c-38cc0639f7ee"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.5/7.5 MB\u001b[0m \u001b[31m37.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m268.8/268.8 kB\u001b[0m \u001b[31m25.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m107.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m80.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##3. Understanding Embeddings\n",
        "Embeddings transform categorical variables into continuous vector space. In the context of NLP, words or phrases are mapped to vectors of real numbers. The goal is to have semantically similar items represented by vectors that are close in this space.\n",
        "\n",
        "##4. Deep Dive into Token Embeddings\n",
        "Token embeddings convert individual words or sub-words into a dense vector format."
      ],
      "metadata": {
        "id": "Nn8PfHylwLn_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import BertTokenizer, BertModel\n",
        "\n",
        "tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\n",
        "model = BertModel.from_pretrained('bert-base-uncased')\n",
        "\n",
        "input_text = \"The world of embeddings is vast.\"\n",
        "tokens = tokenizer(input_text, return_tensors='pt')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145,
          "referenced_widgets": [
            "85962e48f2d2442cba4151bd6930ae9c",
            "8dd9f660ef2e44b9bfa7a6261ac08ba7",
            "4794a0a52f564f13ae3882c3b3060e8d",
            "0d90e6d600284ee597f7d6085f5fafb8",
            "e453a5aa43f7442f8f36c9609c0c0e8c",
            "2cfeb1677ec24e078a67b88a68a0f417",
            "ef0b8c38eda24ef2beb4fafab3803947",
            "d686d880344a4451bbec22f5d59933c2",
            "573e47ae4244410bb7d94e9254765e9b",
            "c1a998ccdae948158944404c564e4c39",
            "43d7325199d2458da0809f72282c5d9e",
            "2b94ad1e89e048f0a6121eb352f1db7f",
            "9a2dea225e6845dfafa19e45aaf5dbfe",
            "3862e7c4587f412981e5942a9dc9f6cb",
            "346f11c6031641b082a109f2c45da0d3",
            "bec22b9870674282af9797929ea154c3",
            "b9d38c87444949c399206f644804f6fd",
            "696a97128cc342738076343c298bbee9",
            "26d9d1914b874de1ba68a3eb3881c4a5",
            "8adcf7b7db6b4e679505e848ef651a5f",
            "d8f401d1173c43b4bae3c8c1a3214b84",
            "34e30849de5948208424f4900a9dfffa",
            "cd08738f01a441fea32b06803a12ee1f",
            "2317a7610b444efc901270a2ae5ff32a",
            "f377afae00fa4ec0b5283ef87cd5da9b",
            "a82ae221d0434d8aad909d7309d432ad",
            "169a5aea8df04a66a2202fe329dd8d2a",
            "8995160597e540628dc4fd94420d431b",
            "a2450d05a87e4fcea27e5aca1dd7cf3f",
            "12f72f729d7a410788e58e0acce5c3ff",
            "51ddb379365c4aae9414afee809816c5",
            "d9ef77507a134c25b7368087c84974ff",
            "0570e6d2f3594afa8851e1ff0ff816c2",
            "a859d97a8bce49d7b32b18f38e58dfe7",
            "f2937a141b39436dadf80f8a1767e239",
            "cf3b47a516db4c728212891b4bd0a1f3",
            "b9ac95eff849486c805af6d358e019eb",
            "a156cd35bfb94ec48b21ddb99341bc77",
            "17c22bc6de5b486abcb312289830d4b0",
            "fc9459527210450c8e00301f37384ee9",
            "e176ff18d3764443bf200b72df1a7340",
            "c8dbcd3d362d48f89c97f40b84b86d65",
            "0af689241e7848a780158989c68ab7aa",
            "1233f7a9d7e642799a0307187ec64b2d"
          ]
        },
        "id": "U9_zYb_QwJi_",
        "outputId": "5cbc1c4e-24a7-4445-bbd6-17945cf3b0d5"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)solve/main/vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "85962e48f2d2442cba4151bd6930ae9c"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)okenizer_config.json:   0%|          | 0.00/28.0 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "2b94ad1e89e048f0a6121eb352f1db7f"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)lve/main/config.json:   0%|          | 0.00/570 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cd08738f01a441fea32b06803a12ee1f"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading model.safetensors:   0%|          | 0.00/440M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "a859d97a8bce49d7b32b18f38e58dfe7"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This code uses BERT, a popular transformer model, to generate token embeddings.\n",
        "\n",
        "##5. Unraveling Positional Embeddings\n",
        "To grasp sequences or order of words, transformers use positional embeddings. They encode the position of a token in the sequence, compensating for the transformer's architecture which doesn't inherently understand order.\n",
        "\n",
        "##6. Visualizing Embeddings for Clarity\n",
        "Visual aids can demystify embeddings:\n",
        "\n",
        "**Token Embeddings:**\n",
        "Use Principal Component Analysis (PCA) to reduce the dimensionality and visualize:"
      ],
      "metadata": {
        "id": "V8H5-qEFwSqt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn.decomposition import PCA\n",
        "\n",
        "word_embeddings = model(**tokens).last_hidden_state[0].detach().numpy()\n",
        "pca_result = PCA(n_components=2).fit_transform(word_embeddings)\n"
      ],
      "metadata": {
        "id": "3zzRzYmSwPAO"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Plotting this gives a bird's-eye view of how tokens are related.\n",
        "\n",
        "###Positional Embeddings:\n",
        "Extract from the model and visualize:"
      ],
      "metadata": {
        "id": "ALxtWKuewbKo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pos_emb = model.embeddings.position_embeddings.weight.detach().numpy()\n",
        "plt.imshow(pos_emb)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 413
        },
        "id": "jfj9qjaRwXaY",
        "outputId": "35c1afca-dc23-4f90-8a40-2a0a32011adc"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f24cc33b640>"
            ]
          },
          "metadata": {},
          "execution_count": 4
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 7. The Mathematics Behind Embeddings\n",
        "\n",
        "Embeddings utilize high-dimensional spaces to capture semantic relationships. Essentially, embeddings are the result of optimization problems where each word's representation gets refined over time during model training.\n",
        "\n",
        "### 7.1 Vector Representation\n",
        "\n",
        "For a vocabulary \\( V \\) of size \\( N \\), each word can be represented as a one-hot vector of size \\( N \\). If \"apple\" is the \\( i^{th} \\) word in \\( V \\), its one-hot vector \\( o \\) has a 1 at the \\( i^{th} \\) position and 0 elsewhere.\n",
        "\n",
        "$$ o_{apple} = [0, 0, ..., 1, ..., 0, 0] $$\n",
        "\n",
        "### 7.2 Embedding Matrix\n",
        "\n",
        "The embedding matrix \\( E \\) of size \\( N \\times d \\) (where \\( d \\) is the desired embedding dimension) is multiplied by the one-hot vector to produce the dense embedding vector \\( v \\) for the word.\n",
        "\n",
        "$$v_{apple} = E \\times o_{apple}$$\n",
        "\n",
        "### 7.3 Objective Function in Word2Vec\n",
        "\n",
        "For simplicity, let's consider the Skip-Gram model. Given a center word \\( c \\) and a context word \\( o \\) from its surrounding, the objective is to maximize the following probability:\n",
        "\n",
        "$$ P(o|c) = \\frac{exp(u_o^T v_c)}{\\sum_{w=1}^{N} exp(u_w^T v_c)} $$\n",
        "\n",
        "Where:\n",
        "- $ v_c $ is the embedding of the center word.\n",
        "- \\( u_o \\) and \\( u_w \\) are the output vector representations of the context word and any word in the vocabulary, respectively.\n",
        "\n",
        "The goal is to adjust the embeddings to maximize the likelihood of the context word given the center word.\n",
        "\n",
        "### 7.4 Training via Gradient Descent\n",
        "\n",
        "The embeddings are learned by minimizing the negative log likelihood of the objective function:\n",
        "\n",
        "$$ J = -\\log P(o|c) $$\n",
        "\n",
        "This is achieved using gradient descent. The gradient with respect to the embedding vectors and the output vectors is computed and used to update the vectors iteratively.\n",
        "\n",
        "### 7.5 Positional Encoding in Transformers\n",
        "\n",
        "In transformers, positional information is encoded using sinusoidal functions:\n",
        "\n",
        "$$ PE(pos, 2i) = \\sin\\left(\\frac{pos}{10000^{2i/d}}\\right) $$\n",
        "$$ PE(pos, 2i+1) = \\cos\\left(\\frac{pos}{10000^{2i/d}}\\right) $$\n",
        "\n",
        "Where:\n",
        "- \\( pos \\) is the position of the word in the sequence.\n",
        "- \\( i \\) is the dimension.\n",
        "\n",
        "The sinusoidal pattern allows the model to attend to relative positions, and since the pattern is deterministic, it doesn't need to be learned.\n",
        "\n"
      ],
      "metadata": {
        "id": "FjwN4Ev4wfFR"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Segment Embeddings: Beyond Tokens and Position\n",
        "\n",
        "When building natural language processing models, especially those tailored for tasks like question answering, entailment, or sentence-pair classification, it often becomes necessary to feed more than one sequence into a model. In such cases, the model needs to be aware of which tokens belong to which sequence. This is where segment embeddings come into play.\n",
        "\n",
        "### Definition\n",
        "\n",
        "Segment embeddings, often called sentence embeddings or token type embeddings, are vectors used to distinguish different sequences when they are fed into a transformer model together.\n",
        "\n",
        "### How Do They Work?\n",
        "\n",
        "Let's take the BERT (Bidirectional Encoder Representations from Transformers) model as an example. BERT is pre-trained on concatenated pairs of sentences. When two sentences are fed into BERT, the first word of each sentence gets a special token, \\([CLS]\\) and \\([SEP]\\), respectively. Apart from the token and positional embeddings, BERT uses segment embeddings to differentiate between the two sentences. The tokens from the first sentence get a \\textit{segment A} embedding, while the tokens from the second sentence get a \\textit{segment B} embedding.\n",
        "\n",
        "### Mathematical Formulation\n",
        "\n",
        "Given:\n",
        "1. **Token Embeddings** (\\( E_t \\)): This represents the embedding for each token.\n",
        "2. **Positional Embeddings** (\\( E_p \\)): This provides information about the position of the token in the sequence.\n",
        "3. **Segment Embeddings** (\\( E_s \\)): There are usually only two of these (for \\( A \\) and \\( B \\)). Each token in a sequence gets the embedding of its corresponding segment.\n",
        "\n",
        "The final embedding for a token is the sum of these three embeddings:\n",
        "\n",
        "\\[ E_{final} = E_t + E_p + E_s \\]\n",
        "\n",
        "This combined embedding is then used as input to the Transformer.\n",
        "\n",
        "### Why Are Segment Embeddings Important?\n",
        "\n",
        "1. **Sequence Differentiation**: As stated earlier, in tasks where two sequences are fed together, segment embeddings help in distinguishing tokens from different sequences.\n",
        "\n",
        "2. **Model Generalization**: They allow models like BERT to be fine-tuned on downstream tasks that require understanding of two distinct sequences, like question answering where a model needs to differentiate between a question and a passage.\n",
        "\n",
        "3. **Semantic Understanding**: By differentiating sequences, the model can better capture semantic relationships between them. For instance, in sentence-pair classification tasks like entailment, understanding the relationship between two sentences is crucial, and segment embeddings aid in this.\n",
        "\n",
        "### Limitations\n",
        "\n",
        "Segment embeddings in models like BERT are binary; they can differentiate between two sequences (\\( A \\) and \\( B \\)). If there's a need to process more than two sequences simultaneously, this binary embedding approach won't suffice. However, one could theoretically expand this concept to handle more sequences, but that would also require changing the pre-training process to accommodate these additional segments."
      ],
      "metadata": {
        "id": "U3KVClpPxllm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##9. Custom Embeddings: Tailoring to Your Needs\n",
        "For domain-specific applications, training custom embeddings on specialized datasets can provide more relevance.\n",
        "\n",
        "##10. Challenges and Limitations\n",
        "Embeddings, though powerful, aren't free from challenges. They might fail to capture polysemous words effectively, and cultural biases present in data can be perpetuated in embeddings.\n",
        "\n",
        "##11. Future Developments in Embedding Space\n",
        "Research in embedding space is burgeoning. Future directions include dynamic embeddings that change based on context, and zero-shot embeddings that can represent words not seen during training.\n",
        "\n",
        "##12. Conclusion\n",
        "Embeddings are the linchpin of transformer architectures, bridging human language and machine understanding. As we harness their potential and address their limitations, the future of NLP shines even brighter."
      ],
      "metadata": {
        "id": "wa6M9tmpxiQ3"
      }
    }
  ]
}